Episode 133: March 18, 2021
02:00: Governance Team Update
10:02: Smart Contracts Team Update
17:40: Risk Team Update
22:40: Real World Finance Update
29:27: Operational Support Update
34:18: MIPs Update
37:00: March MIPs Governance Poll Bundle Review
01:00:30: Liquidation 2.0 Oracle Risks
1:10:10: Incentives and Vesting Discussion
Agenda and Preamble
- Hello everyone, and welcome to the MakerDAO scientific Governance and Risk meeting #133 taking place on Thursday, March 18th at 1700 UTC. My name is LongForWisdom. I’m the Governance facilitator of MakerDAO. I run these meetings, and I’m joined by a bunch of awesome Maker people. You are all probably a little bit sick of hearing me saying this by now.
- A quick reminder about daylight savings before I dive into my sections; for some people, this has already happened, but it’s coming in the next few weeks for others. This meeting is defined in terms of UTC. If you are in a country with daylight savings, the meeting time will shift by one hour.
- If you have comments or questions, please interrupt. We like hearing from the Community. Please get involved.
- Deeper Discussions happen in the Forum.
- We will briefly go through the executives and polls that we had last week.
- Last week’s executive passed, which brings us the ETH-C vault type, a new vault type using ETH as collateral with a lower stability fee and a higher liquidation ratio. It also brings stability fee adjustments to some of the UNI V2 LPs. The complete list is on the portal. We should see that execute soon.
- This week’s polls are now finished. We had the Optimism Dai Bridge poll, which passed. It confirms that the Community, Maker holders, want the Smart Contracts team to pursue Optimism’s Dai Bridge as described on the Forum.
- We had the monthly MIPs Governance poll pass, which is good to see because there are many juicy MIPs included.
- We saw the poll to increase the
dust parameter for most vault types pass.
- We were supposed to have another
dust parameter poll specifically for ETH-B. However, it didn’t go from the portal because there was an error in the poll that we missed until Wednesday, so we decided to skip it next week. This shouldn’t cause grief because ETH-B is currently maxed out anyway.
- We had other polls to increase ETH-A and WBTC-A maximum debt ceilings. For ETH-A, the poll passed to move the debt ceiling from 2.5 billion to 15 billion. For WBTC-A, the polled passed to move the debt ceiling from 350 million to 750 million. This reflects the fact that people are sick of micromanaging this now that we have the Debt Ceilings Instant Access module. The rough consensus is to use the Stability Fee to manage usage in the future rather than strict limits.
- We have greenlit polls still active for nine potential collateral assets. If you haven’t voted for these, please consider doing so.
- The executive this Friday will include the ETH-A and the WBTC-A maximum debt ceiling increases and the
dust parameter change for those vault types.
- We will also have an executive on Monday, the monthly MIPs executive. It will include the final ratification of the MIPs in March’s Monthly Governance Cycle.
- In terms of the Governance Team itself, we are working on a couple of things. The SourceCred payouts’ funding is transitioning to be paid by GovAlpha, which is the first Governance Core Unit dealing with SourceCred and organizing that transition. A couple of us have been working on MIP0 changes to make it more streamlined, easy to understand, and more accessible. I’ve been working on Liquidations 2.0 parameter documentation. Elihu has been working on the Vision Setting Working group. The MIPs portal work continues; however, not quite ready for showing anything new just yet.
Governance at a Glance
Governance at a Glance - Forum Thread
ICYMI - Announcements
- Highlighted Signals
- Other Active Signals
Smart Contracts Team Update
- We’re expecting to do the address rehearsal with Matt, but we could not do so due to the busy schedule.
- For MIP22, we sat with Centrifuge and found a very good path to shoehorn this into MIP21 format. Changes we need to make are very minimal. We will review that, and there is a chance that MIP22 becomes the MIP21 collateral type. Their entire kit may go to Kovan early next week, which is very exciting.
- Security team review work is happening. Once it gets reviewed, we will release some docs on the roadmap there.
- Yesterday, we had the SC core discussion, and MIP45, which is Liquidations 2.0, has been under development.
- We deployed Liquidations 2.0, our release candidate 1, into Kovan. This is its own sort of bespoke deployment of multi collateral DAI into Kovan. It’s not like a normal Kovan staging environment. I dropped all the addresses for the deployment in the #collateralonboardingupdateschannel. Feel free to check it out. Also, take a look at the config file. We made it easy to iterate and test.
- We changed few things like tail and cusp between collateral so that could see different expirations. Suppose keepers are going to test that out. In that case, they can try different things such as integrations, integration partners, or any UIs. Feel free to start using that or at least test the code.
- We’re going to move on working on the Spell for Kovan as part of our staging environment.
- We’ve worked on Ilk-registry changes for Liquidations 2.0. Quantstamp and PWC finished with their final audit reports. TrailofBits and Gauntlet are still ongoing. Several things came out for the clarifications to the MIP. There are also additions to those audit reports now that they have finished. We want to add that into MIP45 at some point. Long, we will be working with you because it’s an informal submission. We don’t want to change yet, but maybe we can submit a link in the comments to the poll request. There are no substantial changes, just typos, and invariants that have been updated.
- We will put up the bug bounty program at the same time as mainnet deployment.
- I want to bring up that we’re working on an issue this week. Which pushed back Kovan deployment. We’re hoping for Monday. If there are any integrators in here crunched for time, please let us know.
Risk Team Update
- We’re currently writing a report on LINK-A proposed parameters for liquidations 2.0. Chris mentioned specific risks that will be addressed later in regards to liquidations 2.0. We’re evaluating to set some more conservative parameters to mitigate these risks as much as possible. Gauntlet has also finished the report, and they’re planning to publish findings on the forum sometime soon. The proposed parameters for LINK-A should be based on our own analysis results and the one made by Gauntlet. We might make a few adjustments when proposing parameters for LINK, such as, for instance, having a bit less aggressive price curve function because LINK has a high liquidation ratio. Next week, we should have this report with parameters released. I hope we can get enough feedback from other teams in due time.
- The other thing we’re also working on now is a new framework for evaluating UNI-LP risk premiums and their debt ceilings. Nik proposed higher debt ceilings for some UNI LPs. We believe apart from auction risks; those debt ceilings are reasonable when looking at the liquidity of underlying tokens. It is harder to evaluate the risk associated with auctions right now for UNI-LPs. We still haven’t had a more significant liquidation event to see how many keepers truly bid on these auctions when more money is stake. Still, it does make us more comfortable knowing that there were actually no zero bids for UNI-LPs at the same time when certain UI’s had a bug. Despite low participation in these auctions, there was still some activity. Also, some of the UNI LP’s collateralization ratio distribution is very healthy. For DAI-ETH, for instance, the median collateralization ratio is about 300. This is high considering the daily liquidation ratio is only 125, presenting fewer risks and lower liquidation probability. I think we can safely increase debt ceilings on UNI-LPs. We also hope that if we have a more significant liquidation event, we will see a decent amount of keepers bidding by increasing the minimum bid amount.
- Besides actively working on liquidations 2.0s and UNI LPs, there’s also some background work, mostly on optimizing our risk calculations. We’re trying to convert risk files to more robust settings and get calculations for AWS so that metric can be shown publicaly on web dashboards. We’re currently working on optimizing slippage curves which are essential when making decisions on debt ceilings or auctions throughout liquidations 2.0. We started to use a 1inch API, which will probably make an excellent tool to measure on-chain liquidity, becoming more important. That’s all I had for today. If there are any questions, please ask.
Real-World Finance Update
- Centrifuge assets will migrate to use MIP21, interacting with MakerDAO directly. This will bring a lot of velocity to some RWAs. That’s good news for Centrifuge and the team.
- We might have solved another issue we had on RWAs, which is the USD-DAI edging cost. People in the real world don’t want to use Dai. They want to use US dollars because there is always a risk of price variation in Dai. We know this will no longer be the case. However, people still want to be guaranteed that using the PSM brings us stability with USDC. Through USDC, you have stability with USD. There is still some work to be done on this front to see if we should do another PSM for RWAs or anything else to see if it will work well in the future. I think that the USDC PSM will be increased. There should be enough space for the next few months.
- Regarding 6S and RWAs in general, I had a good chat with a lawyer who will help us on the legal front specializing in structured finance. After the meeting, the lawyer was reading the Maker White Paper, which is great to hear. He understood why everything is complicated with Maker. I’m not sure that it was a compliment, but I took it as a positive comment.
- We have bad news concerning analytics. Last week I told you that in June, they would change some widgets being used. There will no longer be any pivot table. Currently, the financials for Maker no longer work because the front-end isn’t working. I will try to see what they can do and other analytics platforms we can use to do that. It’s a similar issue to what Primoz is facing. We’ll have a discussion around this later today. We are working a lot with data, so hang tight.
- We have made progress on Fortunafi. We had a good chat with them and People’s Company for a new round of negotiation. I hope everything will be fine. It could be difficult to negotiate on good terms with them, but everything looks great.
- Frank Cruz: Does the RWAs team see any risks with Dai by losing opportunities? We see new platforms coming on and new stable coins looking to bring RWAs into the blockchain ecosystem. You have Centrifuge waiting, and there are new protocols like Acala coming online pretty soon. Yesterday’s launch of Tether USD made a lot of headlines. Do we see a risk here on losing those opportunities since it takes some time to get things going?
- Sébastien Derivaux: I don’t think there is a direct risk for Dai because Centrifuge is doing more and more business, up to 5 million now and increasing quite quickly. They are still on Dai. Also, other competitors that are tokenizing RWAs are no longer using Dai. We had a discussion with Goldfinch that is onboarding RWAs, and they use USDC. That is not good news for us, but it will only pay at the margin. Another thing that can be important, and maybe we should focus more on, is to use RWAs to gain Dai’s market share. Let me explain that: Many companies in Asia are using USDT for daily transactions. They are not using Dai because USDT was first. However, you cannot get a loan in USDT. Suppose we can find a way to give them loans in Dai, backed by invoices and inventory. In that case, it might bring some visibility for Dai to popularly utilized stable coins. In the U.S., not many people are using Dai or UDSC for daily transactions. They are used mainly for farming, which is not the best use case. However, they are using it for doing cross-border business in Asia. By bringing loans to those businesses, we might increase Dai’s value proposition for them. This idea is still ongoing. Otherwise, I don’t see USDC doing a business line on RWAs or stuff like that. We clearly have competitors that want to bring RWAs and not choosing Dai.
- Frank Cruz: Thank you!
Operational Support Update
Operational Support Weekly Update: March 12-18, 2021
- We’re going to do EUR-DAI for the Know Your MIP call next month in April.
- Regarding collateral onboarding; we’re working on the ETH-C vault and RWF as well. We will have Bancor on a collateral call on April 7th, 2021. Then we will have a solid block for RWF on April 14th, 2021.
- For the core unit updates, we have revamped the notion page used to present the core units and try clearing more of the feedback we’ve received. Feel free to give us more feedback on potential improvements.
- Last week LongForWisdom and his team came for session number 2 for launching pod sessions. Last Friday, we had Sablier for the core unit tools. On Tuesday, we had AccountAble with David, and yesterday we had Smart Contracts core unit with Derek.
- Tomorrow, Seb will be presenting the RWF core unit, which will be voted on very soon. Next Wednesday, Seth will be presenting the content production core unit.
- Regarding core units in RFC, we have the same four. David posted the Governance Communications proposals to RFC. Please give feedback.
- Touching on interesting topics of conversation, we’ve had a blind spot check. We also had a call yesterday with Seth. He started working on things we should cover, map the whole space, and analyze which Core Unit would take over which aspects.
- Maker vesting has been a heated topic these last few days, and we’ve been prepared a post on it.
- LongForWisdom: How does Operational Support think the onboarding on Core Units are going?
- Juan: It depends upon what your metric is. It’s exciting to see that three core units took the first cycle already. It’s going well based on the MKR token holders’ reaction and the feedback we have got. We don’t know how the community reacts to different core units being proposed, especially when moving them forward to the voting area. It would be interesting to see. We haven’t still got any core unit rejections as of yet.
MIPs Update #30